Isolation of crystals of a planar nitronyl nitroxide radical: 2-phenylbenzimidazol-1-yl N,N′-dioxide (PBIDO)

Article abstract of DOI:10.1039/a608399d

A planar nitronyl nitroxide radical, 2-phenylbenzimidazol-1-yl N,N′-dioxide, has been isolated in a stable single crystal form. X-Ray analysis has shown that the radical molecules form a dimeric two-dimensional structure. The dominant magnetic interaction

Full text of DOI:10.1039/a608399d

View Article Online / Journal Homepage / Table of Contents for this issue
Isolation of crystals of a planar nitronyl nitroxide radical: 2-phenylbenzimidazol-1-yl
N,N∞-dioxide (PBIDO)
Yoshio Kusaba,a,b Masafumi Tamura,*a† Yuko Hosokoshi,a‡ Minoru Kinoshita,a§ Hiroshi Sawa,a¶ Reizo Katoa and
Hayao Kobayashib‡
aT he Institute for Solid State Physics, Roppongi, Minato-ku, T okyo 106, Japan
bDepartment of Chemistry, Faculty of Science, T oho University, Miyama 2-2-1, Funabashi, Chiba 274, Japan
A planar nitronyl nitroxide radical, 2-phenylbenzimidazol-1-yl N,N∞-dioxide, has been isolated in a stable single crystal form.
X-Ray analysis has shown that the radical molecules form a dimeric two-dimensional structure. The dominant magnetic
interactions have been found to be antiferromagnetic. Anomalous temperature dependence of susceptibility was observed below
10 K. The EPR study indicates that this is ascribable to spin-multiplet states generated in the crystal.
The study of stable organic radicals has been extended more
and more, particularly in accordance with the recent progress
in molecular magnetism.1 For the control of magnetic inter-
actions and magnetism in the solid state, numerous strategies
have been proposed on the basis of various factors operating
in radical crystals.1
make magnetic measurements, although this compound was
put into the unisolable category in ref. 2. We report here the
preparation, X-ray structure and magnetism of the crystals of
this radical.
The neutral radical crystals studied so far are composed of
molecules with non-planar and flexible substituents, such as
methyl and tert-butyl groups. One of the roles of such substitu-
ents is structural hindrance to decomposition;2 they are
believed to stabilize the radicals chemically. On the other hand,
the existence of such flexible units on the molecule sometimes
gives rise to an unexpected structural phase transition at low
temperature due to the structural degrees of freedom.3
The appreciably high stability of the phenyl nitronyl nitro-
xide radicals,4 which can be explained in terms both of the
conjugation over the two NO groups and of large negative
charges on the terminal oxygen atoms, brings the idea that the
nitronyl nitroxide moiety may still retain its stability even if
the usual tetramethylethylene burdening is replaced by an
aromatic ring. This idea does not sound so hopeless, because
the resultant structure still contains no hydrogen on the a-
carbon atom. The expected structural stiÂness of such a
molecule may reduce the structural instability and may lead
to a well-organized crystal structure; only the rotation around
the CMC bond linking the phenyl and benzimidazole groups
remains flexible. Moreover, the imidazole ring fused with
another aromatic ring is expected to provide a new way to
control the spin density distribution over each molecule in the
solid state. This extension of molecular design would aÂord
new features in intra- and inter-molecular magnetic interactions
and may contribute to a new design of stable radical species
of interest.
Experimental
X-Ray crystallography
X-Ray intensity data at 298 K were collected for a single
crystal with dimensions 0.10×0.10×0.05 mm, using a MAC
Science automated four-circle diÂractometer MXC18 with
graphite monochromatized Cu-Ka radiation by h–2h scans up
to 2h=130°. 1528 independent reflection data satisfying
|F|>3s(|F|) were used for the structural analysis. The structure
was solved by the direct method and refined by full-matrix
least-squares analysis with an anisotropic approximation
except hydrogen atoms. An analytical absorption correction
was made. All the procedures were carried out with the
CRYSTAN program package by MAC Science.d
Powder X-ray diÂraction data were collected using a MAC
Science MXP18 System with a rotating anode generator and
a monochromator of single crystalline graphite for Cu-Ka
radiation.
Static magnetic susceptibility measurements
We have succeeded in isolating such a radical, 2-phenylbenzi-
midazol-1-yl N,N∞-dioxide (1, abbreviated as PBIDO)5–8 in
crystalline form. This molecule consists only of planar units.
This is the first example of the above extension of the repertory
of isolable organic radicals. The crystals were so stable that
we were able to study them by X-ray crystallography and
Magnetic susceptibilities were measured for a polycrystalline
sample from 300 to 1.8 K, by use of a Quantum Design MPMS
SQUID magnetometer. Below 10 K, the applied field was kept
as low as 400 Oe in order to avoid saturation eÂects. The data
were corrected for diamagnetic contributions, by subtracting
the diamagnetic contribution estimated from the curve-fitting
described below.
Present Addresses
† Department of Physics, Faculty of Science, Toho University, Miyama
2-2-1, Funabashi, Chiba 274, Japan (tamura@ph.sci.toho-u.ac.jp).
‡ Institute for Molecular Science, Myodaiji, Okazaki, Aichi 444, Japan.
§ Science University of Tokyo in Yamaguchi, Daigaku-dori 1-1-1,
Onoda, Yamaguchi 756, Japan.
d Atomic coordinates, thermal parameters, and bond lengths and
angles have been deposited at the Cambridge Crystallographic Data
Centre (CCDC). See Information for Authors, J. Mater. Chem., 1997,
Issue 1. Any request to the CCDC for this material should quote the
full literature citation and the reference number 1145/34.
¶ Department of Physics, Faculty of Science, Chiba University,
Yayoicho 1-33, Inage-ku, Chiba 263, Japan.
J. Mater. Chem., 1997, 7(8), 1377–1382
1377

View Article Online
EPR measurements
EPR spectra were recorded using a JEOL JES FE1XG X-
band spectrometer. The samples were placed at the centre of
a cylindrical cavity operating in the TE011 mode. For the low
temperature measurements, a helium-flow type cryostat (Air
Products LTR-3-110) was used to cool the sample to 2.3 K.
Results and Discussion
Preparation of the radical crystal
Scheme 3 Reagents and conditions: i, NH OHΩHCl–MeOH, 25%
The target radical was generated by oxidation of 1-hydroxy-
2-phenylbenzimidazole N3-oxide, 2 (Scheme 1). For the syn-
thesis of 2, various methods have been reported.9–14 Schemes
2 and 3 show the two pathways to the precursor 2 applied in
this work. All the reactions in Schemes 2 and 3 were carried
out by following literature procedures.9,14–18 The only modifi-
cation was that we used sulfuric acid instead of perchloric
acid18 in the coupling reaction of benzaldehyde, 3, and o-
benzoquinone dioxime, 9. Scheme 3 was simpler, in spite of
the relatively low yield of the o-benzoquinone dioxime.
Nitrosobenzene, 7, and benzofuroxan, 8, were commercially
obtained.
2
KOH–H O, 30 min (ref. 18); ii, H SO –EtOH, reflux overnight (ref. 14)
2
2
4
C H N O : C, 69.33; H, 4.03; N, 12.44%. Found: C, 69.04; H,
13
9 2 2
4.09; N, 12.46%.
EPR spectra of CH Cl solutions of the obtained crystals of
2
2
1 recorded under air showed clear 152535251 five-split lines
due to the two equivalent 14N nuclei in the molecule. The
observed coupling constant, a =4.2 Oe, was consistent with
N
those reported previously.5–8
Powder X-ray diÂraction was also examined for the pow-
dered solid of 1. A total of 25 diÂraction peaks were observed
between 5<2h<40°. All these peaks can be indexed in terms
of the lattice parameters of the crystal structure described
below.
The oxidation of 2 was carried out using excess PbO in
2
CH Cl or by aqueous NaIO . However, the latter gave a
2
2
4
sizable amount of by-product, reducing the yield of 1.
Therefore, we obtained 1 for the crystallization and physical
measurements mainly by use of PbO . Prior to the oxidation,
2
Crystal structure
2 was purified by dissolution in a 14% aqueous solution of
NaOH and reprecipitation by addition of 6 mol dm−3 HCl to
The molecular structure of the radical 1 is displayed in Fig. 1.
The benzimidazole skeleton including the terminal oxygen
atoms is almost planar, as well as the phenyl ring. The dihedral
angle between the best planes for the benzo and ONCNO
moieties is 1.06°. Moreover, the dihedral angle between the
best planes for the ONCNO moiety and the phenyl ring is
only 10.29°; the molecule is nearly coplanar overall.
it. To a suspension of 0.50 g of 2 in 100 ml of CH Cl , typically
2
2
30 g of powdered PbO was added and stirred for 10 min to
2
yield a slightly yellowish light-green solution of 1. The residual
mixture was filtered o and was put into another 100 ml of
CH Cl followed by stirring for 10 min. The reaction in CH Cl
2 2
2
2
and filtering were repeated three times, until the filtrate became
almost colourless. All the filtrate was again filtered and eva-
porated to dryness under reduced pressure. The yield was
Our idea that the stability of nitronyl nitroxides requires
little stereochemical hindrance, as mentioned in the
Introduction, has thus been demonstrated. In solution, the
stability of the radical 1 is somewhat reduced when compared
to that of the conventional nitronyl nitroxides with the burden
of the saturated ethylene. However, the isolated crystals of 1
were stable enough to examine the structure and magnetic
properties without any special care. This is a rare example of
a neutral organic radical having a nearly planar structure in a
purely crystalline form.
0.39 g (78%). Overnight reaction with PbO resulted in
2
decomposition of 1.
Single crystals of 1 were obtained by evaporation of a
pentane solution of 1 (30 mg in 100 ml) for ca. 2 h. Rhombic
plate-like dark-brown crystals were collected after washing
with hexane. Other solvents, such as ethanol, diethyl ether,
acetone and benzene, gave rise to partial decomposition of the
radical during evaporation, as noted by white precipitation or
colour change of solution from light green to brown. Calc. for
The crystallographic data are summarized in Table 1. The
crystal belongs to the orthorhombic system with the space
group Pcab. The lattice constants are, a=11.483(1), b=
˚
˚
23.655(2), c=7.7924(6) A, V =2116.6 A3 and Z=8. The final
positional parameters are listed in Table 2. The molecular
arrangement is what is called the k-structure19 in the field of
molecular conductors, i.e. two-dimensional packing of dimers,
spreading over the ac-plane (Fig. 2). All the dimers are crystallo-
graphically equivalent and centrosymmetric. They are formed
by ring-over-ring overlap of benzimidazole units [Fig. 3(a)].
Scheme 1
Scheme 2 Reagents and conditions: i, NH OHΩHCl–EtOH, conc. NaOH, 2 h, then dry ice (ref. 15); ii, Cl , 8  HCl (ref. 16); iii, 14% NaOH,
2
2
15 min (ref. 17); iv, Et O, 30 min (ref. 9)
2
1378
J. Mater. Chem., 1997, 7(8), 1377–1382

View Article Online
Fig. 1 Molecular structure of PBIDO in the crystal. (a) Viewed along
the direction normal to the imidazole plane. Atomic numbering
scheme is also shown. (b) Viewed along the O,O direction.
Table 1 Crystallographic data of PBIDO
Fig. 2 Molecular packing viewed along the b-axis. Broken lines show
˚
formula
C
H N O
the short distances; O(14),N(17) 2.943 and O(14),N(16) 3.166 A.
13
9 2 2
formula mass
crystal size/mm
crystal system
space group
225.20
0.10×0.10×0.05
orthorhombic
Pcab
˚
V /A3
2116.6(4)
11.483(1)
23.655(2)
7.7924(6)
8
˚
a/A
˚
b/A
˚
c/A
Z
D /g cm−3
1.41
c
˚
radiation
Cu-Ka (l=1.54178 A)
scan mode
h–2h
total reflections measured
2151
unique reflections
1778
reflections used (|F |>3s(|F |))
1528
0
0
residuals: R; R a
w
0.048; 0.057
goodness of fit: S
1.63
Fig. 3 (a) Face-to-face overlapping within the dimer. (b) The interdimer
aThe function minimized was sum [w(|F |2−|F |2)2], in which w=
close spacing.
0
c
1/[s(|F |)2].
0
The distance between the best planes of benzimidazole units
Table 2 Positional parameters and equivalent isotropic thermal par-
˚
within the dimer is 3.410 A. This face-to-face packing should
ameters with standard deviations in parentheses
provide sizable antiferromagnetic intradimer coupling due to
overlap integral between the SOMOs. The molecules are nearly
orthogonal between the adjacent dimers [Fig. 3(b)]. Relatively
short intermolecular interatomic distances are found between
the terminal oxygen atom and the imidazole unit. For example,
atom
x
y
z
B(eq)
C (1)
0.1120 (2)
0.1209 (2)
0.2177 (2)
0.2251 (2)
0.1373 (2)
0.0424 (2)
0.0333 (2)
0.1447 (2)
0.1880 (2)
0.1281 (2)
0.0318 (2)
−0.0107 (2)
0.0494 (2)
0.2692 (1)
−0.0518 (1)
0.1824 (1)
0.0303 (1)
0.277 (2)
−0.10374 (8)
−0.16429 (8)
−0.1875 (1)
−0.2450 (1)
−0.2805 (1)
−0.2583 (1)
−0.2007 (1)
−0.00892 (8)
0.04410 (9)
0.0887 (1)
0.3752 (2)
0.4038 (2)
0.4851 (3)
0.5126 (4)
0.4603 (3)
0.3790 (4)
0.3511 (3)
0.3874 (2)
0.4230 (3)
0.3477 (3)
0.2439 (3)
0.2092 (3)
0.2841 (2)
0.5425 (2)
0.1896 (2)
0.4420 (2)
0.2770 (2)
0.517 (4)
0.568 (4)
0.481 (4)
0.342 (4)
0.301 (4)
0.492 (4)
0.361 (4)
0.192 (3)
0.139 (3)
4.0 (1)
4.3 (1)
6.1 (1)
7.0 (1)
6.4 (1)
6.9 (1)
5.9 (1)
4.2 (1)
5.2 (1)
6.0 (1)
5.7 (1)
5.0 (1)
4.1 (1)
6.0 (1)
6.0 (1)
4.3 (1)
4.2 (1)
6.12 (0)
6.97 (0)
6.36 (0)
6.86 (0)
5.93 (0)
5.23 (0)
5.95 (0)
5.73 (0)
4.97 (0)
C (2)
C (3)
˚
the interatomic distances around 3 A are O(14),N(17) 2.943,
C (4)
C (5)
O(14),C(1) 3.037, O(14),N(16) 3.166 and O(14),C(8)
˚
C (6)
3.206 A. As shown by de Panthou et al.20 and by Inoue and
C (7)
Iwamura,21 this kind of contact has possibilities of ferromag-
netic coupling due to the orthogonality between the SOMOs.
However, the conditions for this are rather subtle and it would
be easily masked by the larger antiferromagnetic intradimer
contributions.
C (8)
C (9)
C (10)
C (11)
C (12)
C (13)
O (14)
O (15)
N (16)
N (17)
H (3)
H (4)
H (5)
H (6)
H (7)
H (9)
H (10)
H (11)
H (12)
0.0802 (1)
0.02659 (9)
−0.01736 (8)
−0.06990 (7)
−0.09843 (6)
−0.06285 (6)
−0.07633 (7)
−0.162 (1)
−0.260 (1)
−0.320 (1)
−0.282 (1)
−0.185 (1)
0.050 (1)
Static magnetic susceptibility
The temperature dependence of the paramagnetic susceptibility
(x ) measured for a polycrystalline sample of 1 is depicted in
p
Fig. 4. It exhibited a rounded maximum at ca. 45 K. Below
0.293 (3)
this temperature, x steeply decreased to ca. 10 K. With more
0.144 (2)
p
cooling, x again increased down to 3 K, where another small
−0.022 (2)
−0.033 (2)
0.256 (2)
p
maximum appeared. The x value at the second maximum
p
amounts to approximately a third of that at the first maximum,
0.159 (2)
0.126 (1)
and is close to that at room temperature. As shown in the
−0.010 (2)
−0.077 (2)
0.112 (1)
inset of Fig. 4, x gradually decreased to 1.8 K, indicating that
0.020 (1)
p
the x value extrapolated to T =0 is finite. Such behaviour
p
J. Mater. Chem., 1997, 7(8), 1377–1382
1379

View Article Online
Fig. 5 Log–log plot of x T vs. T
p
Fig. 4 Temperature dependence of the paramagnetic susceptibility.
Solid curve shows the fit to eqn. (3). The inset shows a close-up of
the low temperature region.
because the number of triplet species decreases with tempera-
ture. Therefore, this analysis does not give more than a
qualitative estimate. We can conclude that J /k #−40 K and
1
B
the interdimer interactions are antiferromagnetic whose order
of magnitude is J /k ~−10 K.
2
B
(finite x towards T =0) qualitatively resembles that of a
For the sake of estimating the contribution of the low
p
gapless low-dimensional antiferromagnetic system, e.g. a spin-
temperature component of x , which is responsible for the
p
1/2 uniform chain.22
appearance of the second maximum, the product x T is plotted
p
against temperature in Fig. 5. Since x T is proportional to the
p
This low-temperature behaviour of x , the appearance of a
p
second maximum and the slow decrease below 3 K, seems
square of the so-called eÂective moment, this plot often gives
peculiar and puzzling, although it was reproducible. We sus-
pected that this might be due to some lattice imperfection
introduced with cooling. Therefore, we examined the depen-
dence of the behaviour on cooling rate; we compared the
susceptibility data for a slowly cooled (12 h from 290 to 4 K)
sample and those for a rapidly cooled (several minutes from
290 to 4 K) one. However, both were the same. It is not easy
to attribute the peculiar magnetic behaviour to the eÂect of
lattice imperfections.
insight into magnetic interactions in a crystal. The high tem-
perature x T value is consistent with the paramagnetic spin-
p
1/2 behaviour, which yields x T 0.375 emu K mol−1 with g=
p
2 to the limit T 2. The monotonic decrease in x T with
p
cooling to low temperature again indicates that sizeable anti-
ferromagnetic coupling is operating. At ca. 10 K, where the
peculiar behaviour was observed, the x T value is ca.
p
7×10−3 emu K mol−1, ca. a fiftieth of the high temperature
value. In other words, the number of spins responsible for the
The appearance of the first larger x maximum indicates
peculiar behaviour is less than a fiftieth of the total number.
p
that the dominant intermolecular interactions are antiferro-
This is consistent with the above analysis of x , where
p
magnetic. It is most likely that the distinct dimeric structure
having face-to-face overlap gives rise to such a large antiferro-
magnetic coupling. If the intradimer coupling only is relevant,
the system reduces to a simple singlet–triplet model, eqn. (1),
C∞/C#1/50 was obtained. It is almost not possible to derive a
physical model explaining such a fact from the highly sym-
metric and simple molecular packing at room temperature. At
present, we can only point out that one possible origin may
be the eÂect of crystal surface or edge, long-range lattice
modulation, or systematically introduced dislocations.
H=2J ∑ S ΩS ∞
(1)
1
j
j
j
where J (<0) denotes the intradimer exchange coupling, and
1
S and S ∞ stand for the spin-1/2 operators for the molecules
Solid state EPR spectra
j
j
within the j th dimer. The model shows the paramagnetic
Single crystals of 1 exhibit an exchange-narrowed single EPR
line at room temperature. The principal g-values for the static
susceptibility as given by eqn. (2),
x
=(C/T )3/[3+e−2J /k T]
(2)
1 B
field (H) applied along the three crystal axes are, g =2.0083,
p,dimer
a
g =2.0059 and g =2.0057. This is consistent with the orien-
with C=0.50 emu K mol−1 for g=2. This gives a broad
b
c
tation of the ONCNO moieties in the crystal; it is known that
most spin density is concentrated on the ONCNO moiety in
the nitronyl nitroxides. Therefore, no drastic change of this
character occurs in our case. The largest peak-to-peak line-
maximum at T =−1.247J /k where x
=0.151/(J /k )
1
B
p,dimer
1 B
emu mol−1. Qualitative agreement of the observed x with
p
this model is found. However, a satisfactory fit over a wide
temperature range could not be obtained by this model; the
width, DH =3.8 Oe, was observed when Hdb, the normal
x
value at the peak exceeds the observed x value.
pp
p,dimer
p
direction to the ac-plane. With tilting the field toward the
Therefore, interactions other than intradimer ones should be
in-plane directions, a or c, the line became narrower, show-
ing a minimum near magic angle (Fig. 6). Such behaviour is
characteristic of a two-dimensional spin arrangement.
considered to account for the temperature dependence of x .
p
A better fit was obtained when we used the formula for x
p
in eqn. (3),
In order to probe the low temperature magnetism micro-
x =x
/[1−2ꢀzJ ꢁx
/N g2m 2]+C∞/(T −h∞) (3)
p
p,dimer
2
p,dimer
A
B
where the first term corresponds to the major contribution
showing the larger maximum and the second term represents
the tail of the low temperature component. The first term is
the first-order correction of x
for the mean-field ꢀzJ ꢁ
p,dimer
2
stemming from the interdimer interactions, J , where z is the
2
number of nearest-neighbour dimers and the other symbols
have their usual meanings. We obtained C=0.48 emu K mol−1,
J /k =−38 K,
ꢀzJ ꢁ/k =−17 K,
C∞=1.3×10−2 emu
1
B
2
B
K mol−1 and h∞=6.5 K. However, this is not entirely justified
because ꢀzJ ꢁ is comparable to J ; it is not appropriate to
2
1
regard ꢀzJ ꢁ as a simple mean field in such a case. Moreover,
2
the mean field becomes meaningless at lower temperatures
Fig. 6 Angle dependence of the EPR linewidth at 290 K
1380
J. Mater. Chem., 1997, 7(8), 1377–1382

View Article Online
scopically, the temperature dependence of the results was EPR
investigated. Fig. 7 shows the temperature dependence of spin
Hda spectrum is a measure of twice zero-field splitting param-
eter, i.e. 2D ca. 60 Oe. In terms of the point–dipole approxi-
mation, this corresponds to the dipolar coupling between two
susceptibility (x ) obtained from the EPR spectra. The
spin
˚
susceptibilities are basically isotropic and show temperature
spin-1/2 species spaced by ca. 10 A. This is close to the lattice
˚
dependence similar to that by static measurements. The tem-
perature dependence of the g-value was very weak for each
field directions.
period a=11.483 A: the distance between the second-nearest-
neighbouring dimers. It is therefore likely that the multiplet
state is formed over the dimers along the a-axis, although its
origin and detailed mechanism are still in question. More
detailed analyses, as well as other physical measurements, are
now underway.
Below ca. 10 K, the EPR line broadens, and fine structure
satellites appear below ca. 4 K as shown in Fig. 8. The splitting
width of the fine structure is largest in the Hda spectrum; it
amounts to ca. 60 Oe. The width was smallest when the field
is tilted from the a-direction by ca. 60°, indicating that the
dipolar interaction responsible for the fine structure links the
Conclusion
spins placed along the a-axis. In addition to this, Dm =2
It has been demonstrated that the planar organic radical
PBIDO can be isolated as a suÃciently stable single crystal,
although its stability in solution is somewhat lower than those
of conventional nitronyl nitroxides. The successful crystalliz-
ation of this planar radical would expand the possible range
of design directed to molecular magnetism. The PBIDO mol-
ecules are arranged in a dimeric fashion and form a layered
structure in the crystal. The dominant magnetic interactions
are intradimer antiferromagnetic, i.e. J /k ca. −30 K, while
s
transition near H=1600 Oe became observable below 4 K
(Fig. 9). These features indicate that spin-multiplet states are
formed in the crystal at low temperature. It is also found that
the additional enhancement of susceptibility below 10 K comes
from the growing contributions of the fine structure. The
decrease in x and x
states. The largest splitting width of the fine structure for the
below 3 K indicates that some weak
p
spin
antiferromagnetic coupling is operating over the multiplet
1
B
the weaker antiferromagnetic interactions, J /k ca. −10 K,
2
B
are operative between the dimers. Peculiar behaviour of x ,
p
the second small maximum, was observed below 10 K. Only
1/50 spins are responsible for this anomaly. The EPR studies
have revealed that this is related to the appearance of spin-
multiplet states at low temperature. The physical mechanism
behind the low temperature behaviour is in question. Any
model applied to this should explain the following features:
(i) the 1/50 spin number, (ii) appearance of the EPR fine
structure and (iii) gapless susceptibility behaviour at very low
temperature.
We are indebted to Dr Akihiko Hayashi and Professor Yutaka
Ueda for their help with the powder X-ray measurements. We
would like to thank Dr Daisuke Shiomi and Mr Kiyokazu
Nozawa for valuable discussions and experimental support.
This work is partly supported by the Grant-in-Aid for Scientific
Research on Priority Area ‘Molecular Magnetism’ (Area
No. 228/04242103) from the Ministry of Education, Science
and Culture, Japan. Financial supports from the New Energy
and Industrial Technology Development Organization
(NEDO) and Toyota Physical and Chemical Research Institute
are also acknowledged.
Fig. 7 Temperature dependence of the EPR intensity (spin susceptibil-
ity) for (q) Hda, (a) Hdb
References
1
For example, see: D. Gatteschi, Adv. Mater., 1994, 6, 635;
M. Kinoshita, Jpn. J. Appl. Phys., 1994, 33, 5718; J. S. Miller and
A. J. Epstein, Angew. Chem., Int. Ed. Engl., 1994, 33, 385. See also,
Proc. 4th Int. Conf. Molecule-Based Magnets, Salt L ake City, 1994,
ed. J. S. Miller and A. J. Epstein, Mol. Cryst. L iq. Cryst., 1995,
271–274; K. Togashi, R. Imachi, K. Tomioka, H. Tsuboi, T. Ishida,
T. Nogami, N. Takeda and M. Ishikawa, Bull. Chem. Soc. Jpn.,
1996, 69, 2821, and references cited therein.
Fig. 8 Low temperature EPR spectra showing fine structure. The
strong central lines are omitted for clarity.
2
3
J. F. W. Keana, Chem. Rev., 1978, 78, 37, and references cited
therein.
D. Shiomi, M. Tamura, H. Aruga Katori, T. Goto, A. Hayashi,
Y. Ueda, H. Sawa, R. Kato and M. Kinoshita, J. Mater. Chem.,
1994, 4, 1915; Y. Hosokoshi, M. Tamura, M. Kinoshita, H. Sawa,
R. Kato, Y. Fujiwara and Y. Ueda, J. Mater. Chem., 1994, 4, 1219.
E. F. Ullman, J. H. Osiecki, D. G. B. Boocock and R. Darcy, J. Am.
Chem. Soc., 1972, 94, 7049.
4
5
6
7
8
9
A. T. Balaban, P. J. Halls and A. R. Katritzky, Chem. Ind., 1968,
651.
H. G. Aurich and W. Weiss, Chem. Ber., 1973, 106, 2408.
H. G. Aurich and K. Stork, Chem. Ber., 1975, 108, 2764.
T. E. Gough and R. Puzic, J. Magn. Reson., 1976, 23, 31.
F. Minisci, R. Galli and A. Quilico, T etrahedron L ett., 1963, 785.
10 A. J. Boulton, A. C. G. Gray and A. R. Katritzky, J. Chem. Soc.
Fig. 9 The Dm =2 resonance observed at 4 K
(B), 1968, 911.
s
J. Mater. Chem., 1997, 7(8), 1377–1382
1381

View Article Online
11 D. W. S. Latham, O. Meth-Cohn and H. Suschitzky, J. Chem. Soc.,
Chem. Commun., 1972, 1040.
19 For example, see: R. Kato, H. Kobayashi, A. Kobayashi,
S. Moriyama, Y. Nishio, K. Kajita and W. Sasaki, Chem. L ett.,
1987, 507; H. Urayama, H. Yamochi, S. Sato, A. Kawamoto,
J. Tanaka, T. Mori, Y. Maruyama and H. Inokuchi, Chem. L ett.,
1988, 463; D. Jung, M. Evain, J. J. Novoa, M.-H. Whangbo,
M. A. Beno, A. M. Kini, A. J. Schultz, J. M. Williams and
P. J. Nigrey, Inorg. Chem., 1989, 28, 4516.
12 D. W. S. Latham, O. Meth-Cohn, H. Suschitzky and J. A. L.
Herbert, J. Chem. Soc., Perkin T rans. 1, 1977, 470.
13 H. N. Borah, R. C. Boruah and J. S. Sandhu, Heterocycles, 1985,
23, 1625.
14 F. Patzold, H.-J. Niclas and E. Grundemann, J. Prakt. Chem.,
¨
¨
1990, 332, 345.
20 F. L. de Panthou, D. Luneau, L. Laugier and P. Rey, J. Am. Chem.
Soc., 1993, 115, 9095.
15 J. S. Buck and W. S. Ide, Org. Synth., 1943, Coll. Vol. II, 622.
16 O. Piloty and H. Steinbeck, Berichte, 1902, 35, 3112.
17 G. W. Perold, A. P. Steyn and F. V. K. von Reiche, J. Am. Chem.
Soc., 1957, 79, 462.
21 K. Inoue and H. Iwamura, Chem. Phys. L ett., 1993, 207, 551.
22 J. C. Bonner and M. E. Fisher, Phys. Rev., 1964, 135, A640.
18 T. Zincke and P. Schwarz, Annalen, 1899, 307, 28.
Paper 6/08399D; Received 16th December, 1996
1382
J. Mater. Chem., 1997, 7(8), 1377–1382